Maitreya - Assignment 02
Maitreya - Assignment 02
Maitreya - Assignment 02
The EBSD image is composed of pixels. A Kikuchi band observed in the image is also composed of
numerous pixels. As seen above the Kikuchi bands are brighter than their background, and have edges
that are composed of white pixels. To identify a Kikuchi band, these pixels need to be identified first.
For a computer program it is easier to identify a point than a line. Thus a special technique is used to
convert the line into a point, this technique is called “HOUGH Transformation”. This technique was
invented by Paul Hough in 1962 to extract edge features from an image. It is a mapping function that
can convert a point in the image space to a line or curve in Hough space.
Hough transform is a mathematical operation that can represent a line as a point and can identify a line
in real space from a point in transform space. In a traditional coordinate system we can use the equation
𝑦 = 𝑚𝑥 + 𝑐 to represent a line. If we were to use a polar coordinate system with parameters p and 𝜃,
the line equation would become,
𝑝 = 𝑥𝑐𝑜𝑠𝜃 + 𝑦𝑠𝑖𝑛𝜃
Where, p is the distance from origin to line
θ is the angle between p and x direction
Fig 2: - (a, b) Illustration of a Hough transformation. A line in the (x, y) space become a point in the (θ, ρ) space. The points
in a line of (x, y) space become lines in the (θ, ρ) space with a single intersection point which represents the line.
In (θ, p) space the equation is represented by a single point (θ, p). This means that a line in the (x, y)
coordinate system would become a point in the (θ, p) coordinate system, and a point in the (x, y)
coordinate system would be represented as a sinusoidal line in the (θ, p) coordinate system. In the same
way, numerous points belonging to one line in (x, y) coordinate system would be presented as a number
of lines in the (θ, p) coordinate system, all passing through same point. Thus by transforming pixels
formed on the EBSD image to lines and by identifying intersection point, the Kikuchi pattern can be
recognized.
Que 2. Explain about the detectors used in EBSD
Ans.: - The EBSD can be detected using 2 types of detectors, i.e. Indirect Electron Detector and Direct
Electron Detector. There are two types of indirect electron detectors which uses either a charge coupled
device (CCD) or a complementary metal oxide semiconductor (CMOS) sensor.
For obtaining a good EBSD pattern, the SEM has to operate at an accelerating voltage of 10-30 kV, and
a high beam current 0.5-10 nA and an electron probe diameter of 0.02-0.5 μm. EBSD signals also
possess a low signal to noise ratio, because of this the detector has to be very sensitive.
1) Indirect electron detector: -
This is a type of detector in which the electrons that are impinged on the detector screen are
converted into photons using a scintillator (phosphorous screen) and these photons are focused
onto an imaging sensor which could be either a CCD or CMOS sensor. These photons are then
captured by this imaging sensor (resulting in an image) and is transferred to a computer for
subsequent processing and analysis.
In order to obtain an EBSD pattern the specimen must be tilted at an angle that is ~70º with
respect to the primary electron beam. The detector used in this is a photosensitive detector, a
phosphorous screen is used for this purpose. It is kept parallel to the primary electron beam and
at a distance of approximately 20 mm from the specimen. The CCD or CMOS sensor is kept
alongside the phosphorous screen which records the EBSD pattern digitally.
Most EBSD detectors are using CCD sensors, in which the signal that is generated is funneled
through a single charge to voltage output node and is then converted from an analogue to digital
signal, whereas in CMOS sensor the charge to voltage conversion occurs within each pixel and
the whole row of pixels is simultaneously addressed for analogue to digital conversion. High
signal to noise ratio is also retained.
Indirect electrons work on the principle of “Charge Integration”. In this the output is formed
from an analogue integration of the total charge induced by the electrons that has been incident
on the scintillator. These detectors are not free from system noise, but in the case of high gain
system it is very insignificant, except for very low electron doses. As the output is directly
proportional to the energy of the incident electrons.
2) Direct Electron Detector: -
As the name suggests a direct electron detector would detect electrons directly without any
intermediate step (converting electrons impinged on phosphorous screen to photons). Direct
electron detectors have been used quite successfully before in TEM. They work on the principle
of electron counting.
These detectors have a very high electron gain that means the signal generated by a single
electron is far greater than the system noise. Also, the fast, on-pixel electronics system are able
to differentiate between the signal and the noise, and produce a digital count of each single
electron event before the next electron electrons’ arrival. The detectors have a low energy
threshold meaning that it would not discriminate between high and low energy electrons. Each
electron that is detected is given equivalent weightage, so that the contribution of all the
electrons would be similar to the energy distributions of all the BSEs.
References: -
[1] Leng, Y. (2013). Materials characterization: introduction to microscopic and spectroscopic methods. John
Wiley & Sons.
[2] Hawkes, P. W., & Spence, J. C. (Eds.). (2007). Science of microscopy (Vol. 1, pp. 130-131). New
York: Springer.